Unlike traditional studies of drug use which have focused either on individuals or communities, our study examines how focus on micro geographic drug hot spots adds to our understanding of drug use and related health problems. The specific aims of the project are: 1) to compare health outcomes (including mental and physical health, drug use, HIV and STI risk) and antisocial behavior of residents who live in drug hot spots to residents in predatory crime hot spots and non-hot spot places; 2) to identify developmental patterns of health outcomes and anti-social behavior of individuals, and the relationship of those patterns to social and ecological changes in characteristics of places; 3) to measure how social and ecological characteristics of places are related to drug hot spots as compared with the comparison conditions; and 4) to identify (and compare) predictors of developmental trends in health outcomes, drug use and other anti-social behaviors over time in the three study samples. This study would provide significant new knowledge about the causes of drug use and its relationship to health outcomes, including mental and physical health and HIV and STI risk. Such knowledge would lay the foundation for the development of new and innovative intervention strategies that recognize the importance of micro geographic mechanisms in the production of drug use, and related anti- social behavior and health problems. In order to address these issues, a prospective longitudinal study of 450 street segments in Baltimore City, Maryland over a five year period is proposed. The sampling approach involves the selection of 150 drug hotspots as characterized by official data from the Baltimore City Police Department (BPD), and comparison samples of 150 predatory crime hot spots and 150 non-hot spot street segments. These street segments would be drawn randomly from eligible units in the city, and will be selected so as to minimize possible spatial contamination. Four types of data collection will be used to address the research goals of the study: 1) an innovative panel survey (with replacement) conducted three times over a five year period at the street segment level, comprising 4,500 completed surveys per wave; 2) physical observations of characteristics of the street segments conducted simultaneously with the surveys; 3) a qualitative study of a sample of 50 street segments from each of the three study groups; and 4) archival data on characteristics of street segments. Analytic methods in the study include descriptive comparisons of the three groups, hierarchical linear models, longitudinal regression designs, and trajectory analysis.